13197 research outputs found
Sort by
Comparing the Maintainability between Different Code Proficiencies in PyPI Projects
Full text link奈良先端科学技術大学院大学修士（工学）master thesi
コウソク セルソーター ノ タメ ノ フェムトビョウレーザー マルチ パルス ショウシャ ギジュツ ノ カイハツ
Full text link奈良先端科学技術大学院大学修士（工学）master thesi
GenKP: generative knowledge prompts for enhancing large language models
No full textLarge language models (LLMs) have demonstrated extensive capabilities across various natural language processing (NLP) tasks. Knowledge graphs (KGs) harbor vast amounts of facts, furnishing external knowledge for language models. The structured knowledge extracted from KGs must undergo conversion into sentences to align with the input format required by LLMs. Previous research has commonly utilized methods such as triple conversion and template-based conversion. However, sentences converted using existing methods frequently encounter issues such as semantic incoherence, ambiguity, and unnaturalness, which distort the original intent, and deviate the sentences from the facts. Meanwhile, despite the improvement that knowledge-enhanced pre-training and prompt-tuning methods have achieved in small-scale models, they are difficult to implement for LLMs in the absence of computational resources. The advanced comprehension of LLMs facilitates in-context learning (ICL), thereby enhancing their performance without the need for additional training. In this paper, we propose a knowledge prompts generation method, GenKP, which injects knowledge into LLMs by ICL. Compared to inserting triple-conversion or templated-conversion knowledge without selection, GenKP entails generating knowledge samples using LLMs in conjunction with KGs and makes a trade-off of knowledge samples through weighted verification and BM25 ranking, reducing knowledge noise. Experimental results illustrate that incorporating knowledge prompts enhances the performance of LLMs. Furthermore, LLMs augmented with GenKP exhibit superior improvements compared to the methods utilizing triple and template-based knowledge injection.journal articl
Large Language Model Challenges to Detect Cancer-Related Cognitive Impairment from Patient Short Speech
No full textThis study uses language-based Cancer-Related Cognitive Impairment (CRCI) screening to examine the language ability of the participants. This study was conducted to determine whether a natural language processing-based system can detect CRCI or not. We obtained speech samples from participants including patients with cancer and cognitive impairment scores. Using LLMs, we extracted 8 linguistic measurement metrics from the collected data. We divided patients into high-cognitive and low-cognitive) groups. The results did not show any correlation between CRCI and language features derived from participants’ speech using LLMs.conference pape
Efficient Neural Network Implementation for Embedded Devices in Resource-Constrained Environments
Full text link奈良先端科学技術大学院大学博士（工学）doctoral thesi
Process innovations for scalable miniaturization in MOSFETs and micro-concentrator photovoltaic modules
Full text link奈良先端科学技術大学院大学博士（工学）doctoral thesi
Integrated Pre- and Post-Treatment Techniques for Solution-Processed Oxide Semiconductors Towards Low-Temperature Sustainable Electronics
Full text link奈良先端科学技術大学院大学博士（工学）doctoral thesi
Development of a Combination Device of Vibration Tactile Device and Tightening Device to Realize Human-Robot Handover Operation
No full textWith the ever-increasing spread of collaborative robotics, humans and robots working side by side in the workplace have become more common. When working on different subtasks of a bigger main task, the user and robot need only interact on a few occasions, like a handover, instead of being in constant contact. Interacting with the robot becomes a secondary focus for the human and should not distract from the main task. To free the human from having to keep their attention on the robot while ensuring an efficient handover, the use of physical stimuli is suggested. These signals allow the user to understand the robot's state while keeping their other senses free. In this research, we aim to combine two devices: one capable of informing the user about the handover position through vibrations and the other using tightening signals to inform about the state of the gripper. Through this combination, we expect an increase in handover efficiency and better human concentration on the main task. Experiments are performed to compare the completion of a handover task with and without the device.journal articl
Wnt-dependent mechanism of the apical constriction of roof plate cells in developing mouse spinal cord
Full text linkApical constriction of epithelial cells usually occurs in a local portion of epithelial sheet, which results in bending of epithelial tissues. However, it is uncertain whether diffusible signal molecules, like Wnt, regulate such locally restricted events. Here, we show that Wnt ligands are required for apical constriction of Wnt1-expressing roof plate (RP) cells during development of the neural tube. Analysis of Wntless conditional knock-out (cKO) embryos, in which Wnt secretion from Wnt1-expressing roof plate cells is impaired, revealed that RP-derived Wnt ligands are required for phosphorylation of myosin light chain (MLC) and apical constriction of RP cells. Loss- or gain-of-function analysis of β-catenin reveals that this apical constriction is regulated in a β-catenin-dependent manner. Consistent with the timing of apical constriction, Wnt ligands accumulate on the apical side of RP cells. In embryos with Wnt1-expressing RP-specific defects in synthesis of heparan sulfate proteoglycan, apical accumulation of Wnt ligands and apical constriction are impaired. Therefore, we propose that specific accumulation of Wnt ligands on RP cells drives apical constriction of these cells.journal articl